436 Mr. R. S. Willows on the Variation of the 



For a given amalgam two curves could be obtained, which 

 differed entirely in certain important characteristics. Thus, 

 suppose the resistance of a spiral was measured, the bath then 

 heated and then slowly cooled. On again measuring the re- 

 sistance at the same temperature (say 15°) it was found that 

 it was greater after heating than before. This could be re- 

 peated several times, the excess of the final over the initial 

 resistance for a particular heating becoming less and less, until 

 after heating for about six times it was found that a further 

 heating did not increase the resistance ; it was then considerably 

 higher than it was at the same temperature before it had 

 undergone these heatings. This was especially the case if 

 the spiral had been allowed to stand at the temperature of the 

 room for several weeks. When it had reached this con- 

 dition a series of readings at small intervals of temperature 

 was taken and another series was taken as it cooled. From 

 these readings a curve was plotted. The spiral was then 

 allowed to stand at the temperature of the room for several 

 weeks, its resistance being measured every morning at the 

 same temperature. It was found that the resistance gradually 

 fell, slowly for the first three days, most rapidly about the 

 seventh, and then again more slowly. In some cases it took 

 six weeks for the resistance to become steady. When no 

 further fall could be detected during an interval of a week 

 a second cycle was taken, and the curve obtained under 

 these conditions differed entirely from the first. The differ- 

 ence between the initial and final resistance at the same 

 temperature (15°) in the last experiment sometimes amounted 

 to 10 per cent, of the whole initial resistance. The results 

 for typical cases are described below. For convenience in 

 plotting, the resistance in the appended curves is given in 

 arbitrary units. 



Throughout this paper, when two curves are given for a 

 zinc amalgam, that obtained under the first-mentioned con- 

 ditions, i. e. immediately after repeated heatings, so that a 

 further heating did not alter its resistance at 15°, is denoted 

 by A, that obtained after it had been standing some time is 

 denoted by B. 



Fig. 1 shows the curves that were obtained from readings 

 taken on a 4 - 8 per cent, zinc amalgam. 



Curve A. The arrows show whether the temperature was 

 rising or falling. 



It is seen that, starting from 15°, the resistance gradually 

 falls with increasing temperature up to 30°. Between 30° 

 and 45 c it scarcely alters. When the bath cools after being- 

 heated to 100° the resistance falls more rapidly than the 



